Semi-supervised Bibliographic Element Segmentation with Latent Permutations
This paper proposes a semi-supervised bibliographic element segmentation. Our input data is a large scale set of bibliographic references each given as an unsegmented sequence of word tokens. Our problem is to segment each reference into bibliographic elements, e.g. authors, title, journal, pages, etc. We solve this problem with an LDA-like topic model by assigning each word token to a topic so that the word tokens assigned to the same topic refer to the same bibliographic element. Topic assignments should satisfy contiguity constraint, i.e., the constraint that the word tokens assigned to the same topic should be contiguous. Therefore, we proposed a topic model in our preceding work  based on the topic model devised by Chen et al. . Our model extends LDA and realizes unsupervised topic assignments satisfying contiguity constraint. The main contribution of this paper is the proposal of a semi-supervised learning for our proposed model. We assume that at most one third of word tokens are already labeled. In addition, we assume that a few percent of the labels may be incorrect. The experiment showed that our semi-supervised learning improved the unsupervised learning by a large margin and achieved an over 90% segmentation accuracy.
KeywordsHide Markov Model Topic Model Segmentation Accuracy Word Token Topic Assignment
Unable to display preview. Download preview PDF.
- 2.Connan, J., Omlin, C.W.: Bibliography Extraction with Hidden Markov Models. Technical Report US-CS-TR-00-6, University of Stellenbosch (2000)Google Scholar
- 3.Chen, H., Branavan, S.R.K., Barzilay, R., Karger, D.R.: Global Models of Document Structure Using Latent Permutations. In: Proc. of North American Chapter of the Association for Computational Linguistics - Human Language Technologies (NAACL HLT) 2009 Conference, pp. 371–379 (2009)Google Scholar
- 5.Hetzner, E.: A Simple Method for Citation Metadata Extraction Using Hidden Markov Models. In: Proc. of the 8th ACM/IEEE-CS Joint Conference on Digital Libraries, pp. 280–284 (2008)Google Scholar
- 6.Kramer, M., Kaprykowsky, H., Keysers, D., Breuel, T.M.: Bibliographic Meta-Data Extraction Using Probabilistic Finite State Transducers. In: Proc. of the 9th International Conference on Document Analysis and Recognition, pp. 609–613 (2007)Google Scholar
- 7.Lafferty, J.D., McCallum, A., Pereira, F.C.N.: Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data. In: Proc. of the Eighteenth International Conference on Machine Learning, pp. 282–289 (2001)Google Scholar
- 9.Sharifi, M.: Semi-supervised Extraction of Entity Attributes Using Topic Models. Master’s Thesis, Carnegie Mellon University (2009)Google Scholar
- 10.Takasu, A.: Bibliographic Attribute Extraction from Erroneous References Based on a Statistical Model. In: Proc. of the 3rd ACM/IEEE-CS Joint Conference on Digital Libraries, pp. 49–60 (2003)Google Scholar
- 11.Yin, P., Zhang, M., Deng, Z.-H., Yang, D.-Q.: Metadata Extraction from Bibliographies Using Bigram HMM. In: Proc. of the 7th International Conference on Asian Digital Libraries, pp. 1–14 (2004)Google Scholar